The latest update from Spyglass MTG for you;
Spyglass MTG, the specialized consulting firm delivering data, AI, and security solutions on Microsoft platforms, has announced the successful deployment of an AI-powered engineering knowledge platform for FM, one of the world’s largest commercial property insurers. The solution gives more than 1,500 FM field engineers instant access to the standards and technical guidance that underpin the company’s engineering-led approach to risk assessment and loss prevention.
The platform solves the common problem for large organizations of making years of institutional knowledge easy to access when employees need it. For FM, this meant finding a way to unlock almost 200 years of engineering expertise and deliver it in seconds, which is especially helpful for field engineers. The company’s engineering standards fill tens of thousands of pages and cover topics like fire protection and equipment reliability. Because of this, engineers doing risk assessments at manufacturing plants, mines, and other industrial sites often waste time finding very specific information in a library of over 300 PDF documents.
“FM wasn’t looking for a chatbot,” said Dori Albert, CEO of Spyglass MTG. “They were looking for a way to put trusted engineering knowledge in front of experts at the exact moment they needed it. The challenge was preserving the rigor and accuracy those engineers depend on while dramatically reducing the time it takes to find answers.”
Built on Microsoft Azure using Azure OpenAI and Azure AI Search, the platform allows engineers to ask questions in natural language and receive answers grounded in FM’s engineering standards. Rather than organizing information solely by document structure, the system was designed around the way engineers analyze problems and apply technical guidance.
The impact was immediate. During the first two months after launch, FM engineers submitted more than 17,000 queries through the platform. The company estimates each search saves between six and ten minutes, freeing thousands of engineering hours annually for client-facing work. One engineer recently used the platform while preparing for a client presentation during a visit to a mining facility. Faced with only a few minutes to locate the necessary guidance, he was able to retrieve the information, prepare his materials, and confidently present recommendations to the client.
“The combination of Azure OpenAI in Foundry Models and Azure AI Search, together with reasoning built into the prompt engineering, stretched it as far as it can go—to maximum accuracy,” said Toby Denbow, Vice President of Analytics Technology and AI Engineering at FM.
The solution was developed to meet FM’s strict requirements around accuracy and reliability. Engineering recommendations influence underwriting decisions, risk assessments, and client guidance, making trust in the underlying information essential.
Spyglass worked alongside FM and Microsoft to design an architecture capable of delivering rapid responses while maintaining the governance and precision required in a high-stakes environment. The system structures information according to engineering logic and continuously validates results against known, “ground-truth” examples to ensure answers remain reliable as standards evolve. The result is an AI-powered knowledge platform that helps engineers spend less time searching for information and more time applying their expertise.
“Without the partnerships with Microsoft and Spyglass, I don’t think we would have been able to deliver this in the meaningful time that we did,” added Denbow. “Their guidance saved us time and improved reliability.”
The FM engagement reflects a growing trend among enterprises seeking practical, measurable AI deployments that enhance employee productivity and decision-making without compromising governance or trust. Recently featured as a Microsoft case study, the success of the FM deployment has already generated interest from organizations across multiple industries seeking similar approaches to enterprise knowledge management and AI adoption.
For more information about Spyglass MTG and its enterprise AI solutions, visit: https://www.spyglassmtg.com.
Be the first to comment